The stilbene cis-combretastatin A-4 Z-1, isolated from the African bush willow, Combretum caffrum shows exciting potential as an anticancer agent, binding strongly to tubulin and displaying potent and selective toxicity toward tumour vasculature (U.S. Pat. No. 4,996,237, Arizona Board of Regents, Pettit et al, Experimentia, 1989, 45, 209; Lin et al, Mol. Pharmacol., 1988, 34, 200; Grosios et al, Brit. J. Cancer, 1999, 81, 1318; Lin et al, Biochemistry, 1989, 28, 6984; Woods et al, Brit. J. Cancer, 1995, 71, 705; McGown et al, Cancer Chemother. Pharmacol., 1990, 26, 79; El-Zayat et al, Anti-Cancer Drugs, 1993, 4, 19; Dark et al, Cancer Research, 1997, 57, 1829.).

Cis-combretastatin A-4 Z-1 is able to inhibit cell growth at low concentrations (IC50, P388 murine leukaemia cell line 2.6 nM). The potency of trans-combretastatin A-4 E-1 is much lower and inhibits cell growth in the μM range. Arguably, it is the ability of Z-1 and Z-2 to destroy tumour blood vessels, effectively starving tumours of nutrients, which makes them such exciting molecules. Tumour vasculature and the formation of neovasculature were first identified as a target for cancer therapy by Judah Folkman some 30 years ago. The work of Folkman and others has clearly identified angiogenesis and blood supply as necessary requirements for primary tumour growth, invasiveness and metastasis. It is now becoming clear that the selective destruction of tumour vasculature will have a significant impact on the clinical treatment of cancer. Angiogenesis is subject to a complex process of regulation and thereby offers a multitude of molecular targets for drug design.
The use of Z-1 as a clinically useful anticancer agent has been severely hampered by its poor water solubility (Brown et al, J. Chem. Soc., Perkin Trans. 1, 1995, 577). The phosphate salt Z-2 is more soluble in water than Z-1 and is soon to enter phase II clinical trials (Pettit et al, Anti-Cancer Drug Des., 1995, 10, 299). Nevertheless, both Z-1 and Z-2 are not targeted towards cancer cells and their therapeutic efficacy would be improved if their selectivity were better. The low solubility of cis-combretastatin A-4 in water and saline has led to attempts in the art to make related compounds or prodrugs which retain the activity of cis-combretastatin A-4 as an anticancer agent and which have enhanced solubility. These attempts focus on forming salts or derivatives at the phenolic hydroxyl group of combretastatin. By way of example, U.S. Pat. No. 5,561,122 (Arizona Board of Regents) discloses the sodium and potassium salts of cis-combretastatin A-4 and a hemisuccinic acid ester derivative, and WO99/35150 (Arizona Board of Regents) discloses the lithium, caesium, magnesium, calcium, manganese and zinc salts of cis-combretastatin A-4, and ammonium cation salts with imidazole, morpholine, piperazine, piperidine, pyrazole, pyridine, adenosine, cinchonine, glucosamine, quinine, quinidine, tetracycline and verapamil.
At the molecular level, both compounds target tubulin, binding strongly at or close to the colchicine (3) binding site, preventing polymerisation of α,β-tubulin heterodimer to microtubules. Their inhibition of microtubule formation prevents mitosis and is important in disrupting the growth of new vascular epithelial cells. In addition, disruption of the intracellular microtubule networks by combretastatin A4 leads to the destruction of microvessels within the tumour. This antivascular activity offers exciting therapeutic possibilities as the destruction of microvessels results in the death of all tumour cells which depend on the vessel for nutrients and oxygen. The multi-functional role of tubulin in both healthy and cancer cells highlights the need for selectively targeted drugs.
We have previously investigated the tubulin-binding properties of agents related to Z-1 and 3 and as part of this effort, we have designed many related compounds that behave in a similar fashion to Z-1 (Ducki et al, Bioorg. Med. Chem. Lett., 1998, 8, 1051; Zhao et al, Eur. J. Nuc. Medicine, 1999, 26, 231; Aleksandrzak et al, Anti-Cancer Drugs, 1998, 9, 545). However, it remains a problem in the art in designing effective compounds and especially those which can be selectively targeted.